A NOVEL MULTILAYER NEURAL NETWORKS TRAINING ALGORITHM THAT MINIMIZES THE PROBABILITY OF CLASSIFICATION ERROR

被引：24
|
作者：
NEDELJKOVIC, V
机构：
[1] Department of Computational and Applied Mathematics, University of Witwatersrand
来源：
IEEE TRANSACTIONS ON NEURAL NETWORKS | 1993年 / 4卷 / 04期
关键词：
D O I：
10.1109/72.238319
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
In this paper, a new multilayer neural networks training algorithm that minimizes the probability of classification error is proposed. The claim is made that such an algorithm posesses some clear advantages over the standard backpropagation (BP) algorithm. The convergence analysis of the proposed procedure is performed and convergence of the sequence of criterion realizations with probability one is proven. An experimental comparison with the BP algorithm on three artificial pattern recognition problems is also given.
引用
收藏
页码：650 / 659
页数：10
相关论文
共 50 条
  • [21] Training Neural Networks Using Salp Swarm Algorithm for Pattern Classification
    Abusnaina, Ahmed A.
    Ahmad, Sobhi
    Jarrar, Radi
    Mafarja, Majdi
    ICFNDS'18: PROCEEDINGS OF THE 2ND INTERNATIONAL CONFERENCE ON FUTURE NETWORKS AND DISTRIBUTED SYSTEMS, 2018,
  • [22] Training recurrent neural networks by using parallel recursive prediction error algorithm
    Chen, DQ
    Chan, LW
    ICONIP'98: THE FIFTH INTERNATIONAL CONFERENCE ON NEURAL INFORMATION PROCESSING JOINTLY WITH JNNS'98: THE 1998 ANNUAL CONFERENCE OF THE JAPANESE NEURAL NETWORK SOCIETY - PROCEEDINGS, VOLS 1-3, 1998, : 1393 - 1396
  • [23] Dynamic Construction of Multilayer Neural Networks for Classification
    Liu, Jiqian
    Jia, Yunde
    ADVANCES IN NEURAL NETWORKS - ISNN 2011, PT I, 2011, 6675 : 514 - 520
  • [24] An efficient recursive total least squares algorithm for training multilayer feedforward neural networks
    Choi, NJ
    Lim, JS
    Sung, KM
    ADVANCES IN NEURAL NETWORKS - ISNN 2005, PT 1, PROCEEDINGS, 2005, 3496 : 558 - 565
  • [25] An Enhanced Training Algorithm for Multilayer Neural Networks Based on Reference Output of Hidden Layer
    Y. Li
    A. B. Rad
    W. Peng
    Neural Computing & Applications, 1999, 8 : 218 - 225
  • [26] Genetic Algorithm based training for multilayer Discrete-Time Cellular Neural Networks
    López, P
    Vilariño, DL
    Cabello, D
    ENGINEERING APPLICATIONS OF BIO-INSPIRED ARTIFICIAL NEURAL NETWORKS, VOL II, 1999, 1607 : 467 - 476
  • [27] A training algorithm for multilayer neural networks of hard-limiting units with random bias
    Zhu, HB
    Eguchi, K
    Tabata, T
    IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES, 2000, E83A (06) : 1040 - 1048
  • [28] An enhanced training algorithm for multilayer neural networks based on reference output of hidden layer
    Li, Y
    Rad, AB
    Peng, W
    NEURAL COMPUTING & APPLICATIONS, 1999, 8 (03): : 218 - 225
  • [29] Optimal Parameter for the Training of Multilayer Perceptron Neural Networks by Using Hierarchical Genetic Algorithm
    Orozco-Monteagudo, Maykel
    Taboada-Crispi, Alberto
    Gutierrez-Hernandez, Liliana
    INTERNATIONAL ELECTRONIC CONFERENCE ON COMPUTER SCIENCE, 2008, 1060 : 3 - 6
  • [30] A novel multicriteria optimization algorithm for the structure determination of multilayer feedforward neural networks
    Kottathra, K
    Attikiouzel, Y
    JOURNAL OF NETWORK AND COMPUTER APPLICATIONS, 1996, 19 (02) : 135 - 147
12345